HEQ5 / Ekos attempting random meridian flips, or just turning tracking off, unsure of cause.

[pipnina]

I noticed this once or twice before, where I'd wake up having set the rig to work, and realise half my subs were just star trails as tracking had turned off randomly. When I most recently set up my kit, it did this while I was setting up and trying to set the focus on the scope and saw the star drift out of my zoomed in live feed.

Not only that, but once I got it taking pictures, it would randomly decide to perform a meridian flip (seen in Ekos' EQMod log), (pointed at Sadr, so absolutely no need for one) and the first I would notice was PHD telling me it was "insufficient to correct". Not sure if it succeeded at moving the mount for one or not because PHD stopped looping at that error message. I'd park the mount and try again, and sure as the sun rises it happened again. Does the HEQ5 circuity have any of this logic built in or will it ONLY perform movements when instructed by the handset or EQMod driver PC? If the latter, then I can rule out hardware issues and blame either my Raspberry Pi or Ekos for whatever is going on.

I've intentionally avoided system updates to avoid regressions or software bugs being introduced, and it wasn't doing this until recently.

I checked the system time, which was correct, and the slew accuracy and using EQMod control panel to slew to RA/DEC coords sent me to the right places, so time should be right. I added my GPS to 3 decimals taken from google maps in Kstars so in theory that's fine too... Ekos has been a good software setup and IMO more intuitive than N.I.N.A, but it has been riddled with bugs too, anyone else experienced this one?

[vineyard]

Do you recalibrate the mount model every time?  I used to have this - I think its something to do with the mount model vs the EKOS model.  When I started basically re-calibrating the mount model at the start of every session (even though it was on a pier so thankfully no PA required every time) it got better.

Maybe also trying posting a query on indilib.org?

[pipnina]

3 minutes ago, vineyard said:

Do you recalibrate the mount model every time?  I used to have this - I think its something to do with the mount model vs the EKOS model.  When I started basically re-calibrating the mount model at the start of every session (even though it was on a pier so thankfully no PA required every time) it got better.

Maybe also trying posting a query on indilib.org?

Recalibrate? I'm not sure what you mean by that as I thought the mount "forgot" everything when it powered down, and assumed it was in the park position at switch-on?

Is there a specific routine I need to run?

[vineyard]

My memory is a bit hazy on the way the different models work (I think the mount has its own model in there that does not get purged at power down).  So what I used to do was once the mount was on & tracking (& polar aligned if you're not on a pier) then I'd purge the model (go to the mount tab and there's a "Clear Model" button).  Then go to the Align tab (the one that looks like a bullseye - I think its called Align), and there's a button you can click to set up a mount model.  I'd suggest choosing 3 named stars and once its gone through that, it has a mount model as of that night's setup, and you're hopefully more likely to get flips to happen when they should.

If that doesn't fix it, I'd try & post a query on indilib.org too.

Edited May 24, 2022 by vineyard